George J. Kaloyanides

Effect of gentamicin on lipid peroxidation in rat renal cortex.

We examined the hypothesis that lipid peroxidation participates in the pathogenesis of aminoglycoside-induced nephrotoxicity. Male Sprague-Dawley rats were injected subcutaneously with gentamicin, 100 mg/kg per day, for 1-4 days. Twenty-four or forty-eight hours after the last injection the rats were killed and the renal cortex was processed for total phospholipids, malondialdehyde (MDA), phospholipid fatty acid composition, superoxide dismutase, catalase and glutathione. Gentamicin induced a significant increase in total renal cortical phospholipids which was evident after a single injection and by the third injection reached a plateau 17% above the baseline level. MDA, an end product of lipid peroxidation, increased from 0.674 +/- 0.021 nmole/mg protein in the control group to 0.931 +/- 0.053 nmole/mg protein (P less than 0.001) 48 hr after the fourth injection. As another index of lipid peroxidation, we determined the shift from polyunsaturated to saturated fatty acids of renal cortical phospholipids. By the second injection of gentamicin we detected a significant decline of arachidonic acid (20:4) present in phospholipid. By the fourth injection, arachidonic acid had fallen 48% below control and was accompanied by reciprocal increases of more saturated fatty acids including linoleic (18:2), oleic (18:1) and palmitic (16:0) acids. The number of double bonds per mole of fatty acid declined from a baseline value of 1.62 +/- 0.01 to 1.20 +/- 0.02 (P less than 0.001) by the fourth injection of drug. Superoxide dismutase showed no consistent alteration, whereas catalase activity (k) fell from the control value of 0.221 +/- 0.007 min to 0.155 +/- 0.009 min (P less than 0.01) by the third injection, where k is the first-order rate constant. Total and reduced glutathione declined after the fourth injection of gentamicin accompanied by a shift to oxidized glutathione with an increase in the ratio of oxidized to total glutathione. These data support the conclusion that accelerated lipid peroxidation occurs early in the course of gentamicin administration and raise the possibility that lipid peroxidation is a proximal event in the injury cascade of gentamicin nephrotoxicity.

Aminoglycoside-induced functional and biochemical defects in the renal cortex

The first site of aminoglycoside-cell interaction occurs at the plasma membrane of renal proximal tubular cells which have been shown to selectively transport and accumulate these drugs. Depression of apical membrane transport of organic base, low-molecular-weight protein, and glucose, together with loss of brush border membrane enzymes and phospholipids in the urine which results in altered phospholipid composition of this membrane, occurs early in the course of aminoglycoside administration. Less well appreciated are the alterations which occur at the basolateral membrane. These include decreased transport of organic bases, Ca2+, Na2+, and K+; increased organic acid transport; decreased activity of Na+-K+ ATPase and adenylate cyclase; decreased calcium content; and altered phospholipid composition. Many of these changes are evident within 90 min of a single injection of drug. Lysosomal dysfunction is manifested by the accumulation of phospholipids in the form of myeloid bodies consequent to the inhibition of lysosomal phospholipases by aminoglycosides. Labilization of lysosomes in vivo has been postulated to be a mechanism of cell injury. Mitochondrial dysfunction attributed to aminoglycosides includes impaired respiration, inhibition of Mg2+ binding, inhibition of Ca2+ uptake, increased permeability to monovalent cations, decreased ammoniagenesis, and decreased gluconeogenesis. However, it remains unclear how the drug gains access to mitochondria in vivo in order to initiate the functional derangements. It is evident that aminoglycosides cause multiple metabolic derangements at multiple sites within renal proximal tubular cells. At present the available evidence does not identify which, if any, of these drug effects is responsible for initiating the injury cascade. The strong possibility exists that aminoglycoside nephrotoxicity reflects the net impact of multiple minor metabolic derangements which individually are of little significance but when added together seriously compromise the cells ability to maintain its structural and functional integrity.

Failure of inhibition of lipid peroxidation by vitamin E to protect against gentamicin nephrotoxicity in the rat.

We tested the hypothesis that accelerated lipid peroxidation, possibly at the level of the lysosome, is linked causally to the pathogenesis of aminoglycoside nephrotoxicity by investigating whether administration of vitamin E would inhibit lipid peroxidation and prevent or ameliorate gentamicin-induced proximal tubular cell injury. Five groups of rats were injected with either saline, vitamin E (600 mg/kg per day) for 6 days, gentamicin (100 mg/kg per day) for 6 days, vitamin E for 6 days plus gentamicin for 6 days or vitamin E for 12 days and gentamicin for the last 6 days. Gentamicin alone induced a 16% increase in renal cortical phospholipids; vitamin E had no significant effect on this change. Gentamicin alone caused accelerated lipid peroxidation evident by a doubling of renal cortical malondialdehyde to 1.23 nmol/mg protein, and a sharp decline of esterified polyunsaturated fatty acids, especially arachidonic acid which fell 43%. These changes were accompanied by depressions of superoxide dismutase, catalase, and total glutathione and a shift from reduced to oxidized glutathione. Concurrent treatment of rats with vitamin E plus gentamicin for 6 days had no significant effect on the gentamicin-induced alterations of malondialdehyde, superoxide dismutase, catalase or the glutathione cascade; however, the shift from polyunsaturated to saturated fatty acids was largely reversed. In rats pretreated with vitamin E for 6 days, gentamicin failed to raise renal cortical malondialdehyde above that of saline-treated rats. The changes in esterified fatty acids were prevented almost entirely, and there were no significant alterations from control of the glutathione cascade. The depressions of superoxide dismutase and of catalase, however, were not reversed. Vitamin E did not affect the amount of gentamicin accumulated in renal cortex nor did it prevent the gentamicin-induced rise of serum creatinine. Examination of renal cortex by light and electron microscopy revealed that vitamin E did not prevent or even reduce the severity of gentamicin-induced proximal tubular cell lesions and necrosis. These results confirm those we obtained in a previous study with the antioxidant diphenyl-phenylenediamine. The observation that inhibition of lipid peroxidation by two distinct antioxidants failed to prevent proximal tubular cell injury and renal dysfunction associated with gentamicin administration leads us to conclude that lipid peroxidation is a consequence and not a cause of gentamicin-induced nephrotoxicity.

Comparison of the Nephrotoxicity of Netilmicin and Gentamicin in Rats

The nephrotoxicity of netilmicin relative to that of gentamicin was examined in Sprague-Dawley rats. Balance studies were performed on rats injected with netilmicin or gentamicin (50 mg/kg per day for 14 days, 100 mg/kg per day for 8 days, and 150 mg/kg per day for 8 days). Control rats were injected with saline. Both drugs caused a dose-related decrease in urine osmolality and increases in urine volume, water intake, and serum creatinine; however, the magnitude of these changes was significantly less in netilmicin- than in gentamicin-injected rats. Light microscopy of renal tissue revealed less proximal tubular cell necrosis in netilmicin- than in gentamicin-injected rats. There was no significant difference between the renal cortical concentrations of the two drugs. Both drugs stimulated uptake of p-aminohippurate in rat renal cortical slices to the same degree. The data indicate that netilmicin is less nephrotoxic than gentamicin in rats, that the difference in nephrotoxicity cannot be explained by a difference in drug concentration in the renal cortex, and that the ability of aminoglycosides to stimulate the organic acid transport system of proximal tubular cells does not correlate with their nephrotoxic potential.

Evidence that the Brain Participates in the Humoral Natriuretic Mechanism of Blood Volume Expansion in the Dog

We examined the role of the central nervous system in the activation of the humoral natriuretic mechanism elicited by blood volume expansion. Studies were performed in anesthetized dogs pretreated with deoxycorticosterone acetate (15 mg/day) and sodium chloride for 12 days. An isolated dog kidney perfused with blood from the femoral artery of the volume expanded dog served as the bioassay system for the humoral natriuretic factor. In group I volume expansion of intact dogs (n = 14) with equilibrated blood promoted an increase in fractional sodium excretion (FE(Na)) from a control level of 2.6+/-0.5 to 13.6+/-1.6%, P <0.001. In the isolated kidney FE(Na) increased from 3.6+/-0.8 to 6.8+/-1.1%, P <0.01. The natriuresis from the isolated kidney occurred in the absence of significant changes in renal arterial pressure, glomerular filtration rate, plasma protein concentration, or packed cell volume, whereas renal blood flow decreased slightly. In group II (n = 20) the dogs were decapitated by means of a specially designed neck vise. In 10 dogs blood pressure was supported by a constant infusion of dopamine (3.8+/-0.7 mug/min per kg body weight). Despite the fact that in response to the same volume stimulus, decapitated dogs manifested an increase in blood volume and cardiac output similar in magnitude to that of intact dogs whereas the rise in mean arterial pressure of decapitated dogs exceeded that of intact dogs, the natriuretic response of decapitated dogs was significantly less than that of intact dogs. FE(Na) in decapitated dogs increased 4.7+/-1.1 compared to 11.1+/-1.4% in intact dogs (p <0.01). Furthermore, volume expansion of decapitated dogs failed to elicit a natriuretic response from the isolated kidney. FE(Na) in the isolated kidney measured 2.6+/-0.4 before and 2.6+/-0.4% after blood volume expansion. These data indicate that decapitation inhibits activation of the humoral natriuretic mechanism elicited by blood volume expansion and are consistent with the interpretation that the brain is the source of the natriuretic factor or that the brain participates in the activation of the humoral natriuretic mechanism at some other site in the body.

Methodology for Study of Isolated Perfused Dog Kidney in Situ

Early in the history of renal physiology, investigators resorted to the isolated blood perfused kidney technique to try and unravel the complex determinants of renal function. The pump perfused isolated kidney technique of Richards and Drinker (1915) and the heart-lung-kidney preparation of Starling and Verney (1925) testify to the high level of sophistication and ingenuity in isolated organ perfusion achieved by investigators of that period. Considering the materials and equipment available, the limited success of these workers in attaining a viable isolated kidney preparation is all the more remarkable. Over the ensuing 50 years, the quest for developing an isolated perfused kidney preparation whose function approximates that of the normal kidney has continued with some, albeit incomplete, success. Nevertheless, the state of the art has reached the point where it is possible, using these techniques, to address certain questions pertaining to renal physiology that cannot be approached or totally resolved by study of the intact kidney. This chapter details three techniques of in situ isolated dog kidney perfusion, the functional characteristics of the preparation, and their experimental application. The term in situ is here broadly defined to include those techniques in which the circulation of the isolated kidney communicates directly or indirectly with the circulation of the dog.

Aminoglycoside Antibiotics Inhibit the Phosphatidylinositol Cascade in Renal Proximal Tubular Cells: Possible Role in Toxicity

A growing body of evidence supports the conclusion that aminoglycoside antibiotics (AG) interact with phosphoinositides. For example AG have been shown to bind to phosphoinositides in model membranes (1–5) by a mechanism best explained by an electrostatic interaction (3,5). The strong avidity of these drugs for phosphatidylinositol-4,5-bisphosphate (PIP2) (1,5,6) has led to the hypothesis that PIP2 serves as the biological receptor for these agents (6–8). AG have been shown to induce a phosphatidylinositol (PI)-enriched phospholipidosis in rat renal cortex (9,10) and in cells grown in culture (11,12), a phenomenon which may be related to the observation that AG have the capacity to inhibit a PI-specific phospholipase C (13–15). Moreover, neomycin has been shown to block the hydrolysis of PIP2 and the generation of inositol trisphosphate (IP3) in response to agonist stimulation in vitro (16,17) and to depress the synthesis and turnover of (32P]PIP2 in vitro and in vivo (1,18,19) . These observations indicate that AG have the potential to perturb the PI cascade, which serves as the transmembrane signal transducing mechanism for a number of agonists (20). Inhibition of the PI cascade by AG might cause profound derangements in the regulation of a number of intracellular processes and thereby contribute to the toxicity of these agents.